Image Classification (CNN using keras)

Image classification is a supervised learning problem: define a set of target classes (objects to identify in images), and train a model to recognize them using labeled example photos.

Neural networks are perfectly suited for image classification: the task of finding the complex patterns in pixels necessary to map an image to its label. As a result, image classification is a common application of deep learning.

To do this, you will be using convolutional layers: layers designed to process image data by focusing on local relationships between features.

watch this video to get an overview of computer vision cases that can be solved using Neural networks, the limitations of neural networks when higher number input features (higher resolution images) are required for training, and how convolutional technique helps in feature extraction and reducing overfitting and improving performance.

Convolution Neural Network (CNN) - Introduction

Follow this Google DL course chapters to understand the concepts involved in creating a CNN.

• Image Classification - Introduction

• CNN

Implementing CNN

For rest of today, follow the best course available on implementing CNN using tensorflow and keras.

You should follow the course and make notes of things you are learning in your own words in your DL-notes.docx file in your Block C Microsoft Teams assignment.

Also, if you are using notebook other than Codecademy session, make sure you upload them to your Block C Github repository.

Codecademy Module : 4.2 Image Classification

Lesson

• 1) Image Classification

Project

• 1) Classifying Galaxies

Quiz

• 1) Classification

Optional - Challenge Project - Covid-19 and Pneumonia Classification with Deep Learning